Attachment of group B streptococci to human embryonic and fetal epithelial cells has been shown to be highly specific and mediated by the poly(glycerolphosphate) backbone of the membrane-anchored lipoteichoic acid (LTA) produced by these organisms. In addition, the role of LTA as a virulence determinant is strongly suggested by the data demonstrating that group B streptococcal isolates obtained from infected infants bind avidly and require high levels of purified LTA to displace them from epithelial cells. In contrast, isolates from asymptomastically colonized infants bind poorly to fetal cells and are easily diplaced by lower levels of LTA. All isolates adhere poorly to adult cells and the bacterial adhesin(s) in this case appears to be group B streptococcal cell surface protein(s). These data indicate that fetal epithelial cells may have receptors that are reduced or absent in older populations and adults. Therefore the present study proposes to purify and characterize the fetal lung receptor(s) that binds the poly(glycerolphosphate) backbone of LTA and compare the binding with the presence or absence of similar receptors present in older individuals (8 days to seven years) and the ability to bind LTA obtained from group B streptococci isolated from infected infants. Since the long chain LTA polymer may serve as a key virulence determinant and a potential diagnostic tool, we propose to generate mutants of group B streptococci through transposon mutagenesis that are deficient in adherence to fetal lung cells and long chain LTA polymer synthesis. The chromosomal region demonstrated to encode the long chain LTA will be cloned in Escherichia coli in order to develop a probe that can be used to examine our large collection of group B streptococcal isolates obtained from infected and asymptomatically colonized infants for a correlation between virulence, adherence, and the presence of the genetic information for long chain LTA synthesis.